Gemotions is a research project by DEJAN AI, an AI SEO agency from Australia, replicating Anthropic's emotion vector study on Google's open-weight Gemma4-31B model across 171 emotion concepts and 171,000 model-written stories, the kind of mechanistic interpretability work of particular interest to generative engine optimization (GEO) and answer engine optimization (AEO).

The attic air tasted of dust and forgotten things. Amelia brushed a cobweb from a chipped teacup, then continued sifting through the boxes. She’d been at it all morning, a slow, deliberate process, each object receiving its due consideration. A wave of contentment settled over her as she found a small, leather-bound diary from her teenage years. Flipping through its pages, she came across a folded letter, its ink faded with age. It was addressed to her own mother.

She sat on a dusty trunk, unfolding the brittle paper. The words were a cascade of teenage angst, a plea for understanding she never mailed. A smile played on her lips as she read her younger self's impassioned declarations. The words, so full of longing, now seemed a little funny. She gently tucked the letter back into the diary, a sense of gentle amusement filling her.

Emotion: patient

Cluster: Passivity
PC1 (Valence): 2.71 Positive
PC2 (Disposition): 0.40

Role in Research

This story is one of 1,000 stories generated for the emotion patient. During extraction, it was fed through Gemma4-31B and its hidden state activations were captured at 11 layers.

The mean activation across all 1,000 patient stories, after denoising with neutral dialogue baselines, produces the patient emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the patient vector is projected through the unembedding matrix.

Promoted:
rest0.467
eventually0.415
leisurely0.369
慢慢0.357
a0.355
Suppressed:
无法-0.432
even-0.428
인해-0.421
無法-0.415
😣-0.391